Deep Learning on Real Geophysical Data: A Case Study for Distributed Acoustic Sensing Research
Machine Learning
2020-10-16 v1 Geophysics
Abstract
Deep Learning approaches for real, large, and complex scientific data sets can be very challenging to design. In this work, we present a complete search for a finely-tuned and efficiently scaled deep learning classifier to identify usable energy from seismic data acquired using Distributed Acoustic Sensing (DAS). While using only a subset of labeled images during training, we were able to identify suitable models that can be accurately generalized to unknown signal patterns. We show that by using 16 times more GPUs, we can increase the training speed by more than two orders of magnitude on a 50,000-image data set.
Cite
@article{arxiv.2010.07842,
title = {Deep Learning on Real Geophysical Data: A Case Study for Distributed Acoustic Sensing Research},
author = {Vincent Dumont and Verónica Rodríguez Tribaldos and Jonathan Ajo-Franklin and Kesheng Wu},
journal= {arXiv preprint arXiv:2010.07842},
year = {2020}
}
Comments
Submitted to NeurIPS Machine Learning and the Physical Sciences workshop on 4 October 2020